It’s an open secret that very few proposed treatments for dementia end up working out, despite extensive and frequently successful tests in mice. So why is mouse dementia so much easier to treat?
According to a new study, one reason might be timing: Researchers tend to begin administering new drugs and other treatments in mice before the animals have clear signs of disease, but human trials almost always focus on people who already have symptoms.
In other words, treatments that show promise in mice are typically preventive, which may explain why they so often fail to help people who are already sick, said Vanessa Langness, a science writer with a background in biomedical sciences working in the lab of Frank Longo, a Wu Tsai Neurosciences Institute affiliate and professor of neurology and neurological sciences at Stanford Medicine.
That conclusion and others emerged from an extensive review of research on tauopathies, a group of disorders linked to Alzheimer’s disease and frontotemporal dementia, conducted by Langness and colleagues in the Longo lab.
“There are currently no approved, disease-modifying treatments for tauopathies, but there are a ton of these preclinical studies (in rodents) that appear to be promising,” Langness said. “We wanted to understand what is limiting the ability of these mouse studies to predict whether a drug would be effective in humans.”
In addition to the timing issue, the researchers identified other potential stumbling blocks for translating preclinical research to human patients: genetic mismatches between mice and people and the fact that most mouse studies focus on only one sex.
The team, which Langness co-led with Danielle Simmons, a senior research scientist in Longo’s lab, reported their results August 18 in a special issue of Alzheimer’s and Dementia.
Taking in the whole field
Although researchers have been trying to treat tauopathies for decades now, nothing has worked and no one really knows why. Langness, Simmons, and the Longo Lab team decided to do something about it. Inspired by a series of annual reports on the state of Alzheimer’s research, the group carried out what’s known as a scoping review: a detailed, systematic look into where a field of study is at, including what’s worked, what hasn’t, and what hasn’t even been tried.
To make their review manageable, they narrowed in on primary tauopathies, in which only tau proteins contribute to neurodegeneration and dementia. That choice, in turn, led the team to focus on preclinical studies in which mice had mutations to a single gene, MAPT, that provides instructions for building tau proteins. When MAPT is mutated, tau proteins sometimes get assembled incorrectly, leading to the tangled protein clumps, or “tangles,” associated with neurodegenerative disease.
Notably, Alzheimer’s is not on the list of primary tauopathies, because the disease involves tau tangles along with other pathological proteins like amyloid beta. Still, “many of the therapeutic strategies we reviewed may also apply to Alzheimer’s,” Langness said, noting that there is growing interest in targeting tau to treat the disease.
But even without Alzheimer’s studies, there was plenty to work with. In all, the team gathered more than 314 research publications covering more than 400 treatment evaluations in mice with MAPT mutations—everything from drugs aimed at preventing tau aggregation and inflammation to treatments involving ultrasound, near-infrared light, and extra virgin olive oil. The studies had evaluated treatments based on a wide range of criteria as well, such as cognitive and motor deficits, neuroimmune responses, and the loss of brain volume.
Your timing is off
As the team sifted through all that research, a few clear patterns jumped out. Most notably, a majority of the MAPT mouse studies (64%) had started treatment either before or at the same time as tangles of tau proteins first started forming in mouse brains—in other words, they tested preventive measures.
That’s not itself a bad thing. Indeed, many scientists believe preventing dementia will be much easier than reversing it. The trouble is, researchers can’t easily test preventive measures for tauopathies in humans. Unlike in Alzheimer’s disease, where biomarkers allow diagnosis and clinical trial enrollment before symptoms emerge, reliable biomarkers for primary tauopathies are still lacking. Although Longo lab scientists and others, like Beth Mormino at the Knight Initiative for Brain Resilience, are working on advanced biomarkers for early disease detection, neuroscientists are still some ways off from reliably identifying who will go on to develop most forms of dementia, much less gathering enough such presymptomatic participants for a well-powered clinical trial.
As a result, Longo said, there’s a mismatch between mouse and human research that could be slowing translational progress in the field. Until scientists are able to test preventive treatments reliably in people with early, presymptomatic disease, he said, more preclinical studies should focus on mimicking the human trials we can do today: testing treatments to halt or reverse the course of more advanced, symptomatic disease
“Just that one statement might influence how people test their mice,” Longo said.
Other factors
Sex was also an issue. Research in human clinical populations has documented significant sex differences in how tauopathies develop and affect cognition—women typically have more difficulty with language, for example, and men tend to exhibit more behavioral changes.
But in preclinical research, “the number of studies that looked at both male and female mice was very low,” said co-lead author Simmons: Only a third of the experiments examined both sexes, while a quarter did not specify what sex they studied. Failing to take note of the differences between male and female mice, Simmons said, could mask real effects that are present in only one sex.
There were other notable observations. For one thing, researchers often injected drugs straight into a mouse’s abdominal cavity, an approach that is rare in humans and could have implications for how the drugs were absorbed.
Another issue was the wide variation in how studies evaluated success. Across studies, researchers reported a dozen different criteria to determine whether a treatment worked—including survival rates, pathological changes in tau, and loss of brain volume—which makes it harder to compare results and identify which approaches might be most effective in humans.
Getting focused
Simply gathering all this information in one place was a significant step, according to Longo.
“Before our paper, no one could even tell you how many strategies had been tried,” he said. “Now we know, and then we can ask what are the main ones, what are the rare ones, what are the patterns—what’s going on out there?”
Knowing what’s going on out there could help the researchers pick out better avenues of exploration, Simmons added. “One of the things we want to do is help the field hone in on promising strategies. “We can ask how many successful mouse trials used a given strategy, and then that might help us make better decisions about which strategies need further preclinical work and which are promising enough to move to humans,” Simmons said.
Among the more promising approaches, Simmons said, were treatments targeting tau aggregation and immune responses; drugs that improved blood flow in the brain or otherwise promoted general neuroprotection; and passive immunization, which involves giving people antibodies to prevent neurodegeneration. But even if those strategies don’t work out, the team is hopeful their study will improve future research.
“Before this, there was no way to take in the whole field,” Longo said. “There are many mouse studies, but it’s hard to look at them in an organized way. We hope our paper can help the field evolve and get more focused.”
Publication Details
Research Team
Study authors were Vanessa Langness, Danielle Simmons, Tyne MacHugh, Robert Butler, James Zhou, Harry Liu, Tao Yang from the Department of neurology and neurological sciences at Stanford Medicine; Tyne MacHugh from the Department of Neurology and Neurological Sciences at Stanford Medicine, the Buck Institute for Research on Aging, the Leonard Davis School of Gerontology at the University of Southern California; and Lisa Ellerby from the Buck Institute for Research on Aging.
Research Support
This work was supported by the Jean Perkins Foundation, Taube Philanthropies, and the National Institute on Aging (T32 AG052374, PO1 AG066591).
Competing Interests
Frank Longo is listed as an inventor on patents relating to a compound, LM11A-31, discussed in the report, that is assigned to the University of North Carolina, University of California, and the Department of Veterans Affairs at San Francisco. Longo is also entitled to royalties distributed by UC and the VA per their standard agreements, and he is a founder, equity holder, board member, and paid consultant for PharmatrophiX Inc., a company focused on the development of small-molecule ligands for neurotrophin receptors that has licensed several of these patents. The remaining authors declare no competing interests.